Home Original page

nested scopes

Robin Becker robin at jessikat.fsnet.co.uk
Sat Feb 3 06:25:23 EST 2001 
In article <mailman.981181863.15223.python-list at python.org>, Tim Peters
<tim.one at home.com> writes
>[Robin Becker]
....
>WRT scopes, the determination of which names are local is entirely done at
>compile-time.  That wasn't always true, but has been true since version
>0.9.9, and was the single biggest speedup in Python's history.  The
>determination of which names are local is still done at compile-time in the
>presence of "import *" and "exec", but then the results can't be trusted,
>and the compiler generates different code that does all sorts of crap under
>the covers to give "the expected" results -- most of the time.  I doubt that
>crap will get extended to live nice with deeply nested scopes, though.
>
>> ...
>> Optimality is the enemy of stability.
>
>in-america-they-tell-us-it's-drugs<wink>-ly y'rs  - tim
>
>
as I suspected this is about optimality; speed vs clarity/simplicity is
the argument here. I guess that the nested scopes stuff has just brought
it to the fore. Luckily we're not forced to use nested scopes in python,
pascal experience makes me despise them because of all that paging up
and down to find the defining declaration. 
-- 
Robin Becker
More information about the Python-list mailing list